• The Transactions of the Korea Information Processing Society
• /
• v.4 no.8
• /
• pp.2106-2119
• /
• 1997

NDAS(Network Data Access System) is a system that interconnects high speed network(B-ISDN) and Local Area Network(Ethernet, FDDI). Nowadays, the request of multimedia data increases rapidly, so NDAS system must accept and process that request efficiently. Comparing with normal data traffic characteristics, multimedia data traffic characteristics are much complex, so it is necessary to implement the system that can cover such factor. In this paper, we suggest a NDAS hardware system model and also we propose a proper buffer size of NDAS system through the simulation. For measuring the proper buffer size of NDAS, we assumed two different simulation models, one is the case of using only one buffer and the other is the case of using two different buffers for Ethernet and FDDI, respectively. After comparing and analyzing simulation results, we suggest the proper NDAS system buffer size.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.1045-1054
• /
• 2020

In this paper, we designed a TCP/IP based ring buffer system that can stably transfer bulk data streams in the unstable network environments. In the scheme we proposed, The observation data stream generated and output by each radio observatory's backend system as a UDP frame is stored as a UDP packet in a large capacity ring buffer via a socket buffer in the client system. Thereafter, for stable transmission to the remote destination, the packets are processed in TCP and transmitted to the socket buffer of server system in the correlation center, which packets are stored in a large capacity ring buffer if there is no problem with the packets. In case of errors such as loss, duplication, and out of order delivery, the packets are retransmitted through TCP flow control, and we guaranteed that the reliability of data arriving at the correlation center. When congestion avoidance occurs due to network performance instability, we also suggest that performance degradation can be minimized by applying parallel streams.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.6 no.6
• /
• pp.383-391
• /
• 2011

As we know the effects of cache memory research, instruction and data caches can be separated for higher performance with Harvard CPUs. In this paper, we shows the efficiency of buffer system in the instruction and data flash storage medium. And we analyzed characteristics of the data and instruction flash and evaluated the performance. Finally, we propose the best buffer structure with an optimal block size and buffer size for the instruction and data flash.

• Journal of the Korean Mathematical Society
• /
• v.40 no.1
• /
• pp.87-107
• /
• 2003

In this paper, we consider a discrete time queueing system fed by a superposition of an ON and OFF source with heavy tail ON periods and geometric OFF periods and a D-BMAP (Discrete Batch Markovian Arrival Process). We study the tail behavior of the queue length distribution and both infinite and finite buffer systems are considered. In the infinite buffer case, we show that the asymptotic tail behavior of the queue length of the system is equivalent to that of the same queueing system with the D-BMAP being replaced by a batch renewal process. In the finite buffer case (of buffer size K), we derive upper and lower bounds of the asymptotic behavior of the loss probability as $K\;\longrightarrow\;\infty$.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.5A
• /
• pp.514-520
• /
• 2004

In WDM networt a WDM buffer is one of essential elements for realizing all-optical packet switching system. In this paper, we propose and demonstrate a single loop type all-optical WDM buffer system. The proposed one consists of NOLM and feedback loop, and provides more than 40 turns buffering (more than 20${\mu}\textrm{s}$) for single input pulse (1550nm, width: 20㎱) before selected by control signal pulse (1553nm, width: 30㎱).

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.10 no.6
• /
• pp.159-164
• /
• 2011

This paper is about the frame for an all terrain vehicle(ATV). An all terrain vehicle, also known as an ATV, has a seat designed to comfortably accommodate a rider over a wide variety of rugged terrain. This study developed and evaluated a buffer system to drive safely and reduce impact force on the load. The Frame includes a main frame portion which has one or more support frames. The buffer system with a double cushion system, one or more of the interaction of the shock absorber is constructed. The 3D modeling of ATV frame is built up with the software CATIA. According to the designing requirement, we analyzed the stress and the deformation of the frame.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.18 no.4
• /
• pp.107-114
• /
• 2010

Coupling systems for trains need more complicated buffer equipments than existing systems because the recent tendency of the regulations enforces trains to be safe for collisions even when the driving speed is higher than before. Using hydraulic buffer is an effective way to satisfy the requirement while it causes the increase of the cost for the coupling system. In this study, we introduce the methodology to build a simulation model for the hydraulic buffer, which could be installed into the coupling systems. In the simulation model of the hydraulic buffer, the reacting force is determined by both buffer stroke and speed whereas the elastic buffer model is designed by using only the buffer stroke in other studies. The simulation results with the advanced hydraulic buffer model shows that the simulating results can be close the real experimental results around 10%, and, if we considers friction forces, the simulation calculates the maximum force within 10% comparing to the experimental.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.6
• /
• pp.1147-1165
• /
• 2011

We present a simple and cost effective rate control scheme for streaming video over a wireless channel by using the information of mobile devices' buffer level. To prevent buffer fullness and emptiness at receivers, the server should be able to adjust sending rate according to receivers' buffer status. We propose methods to adjust sending rate based on the buffer level and discrete derivative of the buffer occupancy. To be compatible with existing network protocols, we provide methods to adjust sending rate by changing the inter-packet delay (IPD) at the server side. At every round-trip time, adjustments of sending rate are made in order to achieve responsiveness to sudden changes of buffer availabilities. A series of simulations and the prototype system showed that the proposed methods did not cause buffer overflows and it can maintain smoother rate control and react to bandwidth changes promptly.

• Proceedings of the KSR Conference
• /
• 2003.10c
• /
• pp.473-478
• /
• 2003

The hydraulic buffer stop placed on the end of the railway brakes the train could not reduce the velocity sufficiently because of the braking system troubles or driver's mistakes. The hydraulic buffer stop is composed of 2 operating parts; hydraulic buffers and rail clamps. Hydraulic buffers brake trains non-destructively in low speed, otherwise rail clamps begin to work in higher speed. In this paper, The braking process of the hydraulic buffer stop is investigated by numerical methods. The hydraulic buffer is numerically analyzed and designed to absorb the kinematic energy of the train below 3.2km/h speed. The hydraulic buffer stop crushed by the train with 5km/h speed is analyzed by FEM package-PAM CRASH in order to obtain the stress profile in rail clamps and buffer stop frame.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.396-398
• /
• 2023

Statista 통계 조사에 따르면 데이터의 규모는 매년 증가할 것으로 예상하고 빅데이터 처리 프레임워크의 관심이 높아지고 있다. 빅데이터 처리 프레임워크 Spark는 Shuffle 과정에서 노드 간 데이터 전송이 일어난다. 이때 분산 처리한 데이터를 네트워크로 전송하기 위해 객체를 바이트 스트림으로 변환하여 메모리 buffer에 담는 직렬화 작업이 필요하다. 그러나 바이트 스트림을 buffer에 담는 과정에서 바이트 스트림의 크기가 메모리 buffer보다 클 경우, 메모리 할당 과정이 추가로 발생하여 전체적이 Spark의 성능 저하로 이어질 수 있다. 이에 본 논문에서는 Spark 환경에서 분산 처리 성능 향상을 위한 직렬화 buffer 최적화 시스템을 제안한다. 제안하는 방법은 Spark Driver가 Executor에게 작업을 할당하기 전 직렬화된 데이터 크기 측정과 직렬화 옵션 설정을 통해 Executor에게 적절한 buffer를 할당할 수 있다. 향후 제안하는 방법의 검증을 위해 실제 Spark 클러스터 환경에서 성능 평가가 필요하다.